Select to view content in your preferred language

Upload a large feature collection

986
4
11-30-2017 01:09 PM
ThomasBjørnson_Larsen
Emerging Contributor

Hi

I want to upload a large dataframe as a feature collection to ArcGis Online. I see that the API has a limit on 1000 features, so do somebody got i tips on how to best split and upload the dataset?

I also want to share the features to the Collector app. Do I have to something more than making sure that the feature collection is shared properly?

Best regards

Thomas

0 Kudos
4 Replies
RandyBurton
MVP Alum

First, check out FAQ: Is there a feature limit in ArcGIS Online?  The limit generally applies as the maximum number of records that will be returned by the server.  Since AGOL is a web service, retrieving too many records can cause the browser to have issues.

When I want to keep the feature small, I will divide the features into areas and create individual maps for each area.  You will need to share the feature with a group, and if the feature is used in a Collector map, the map must also be shared with the group.  You will need to allow Create, Delete, Query, Sync and Update operations for the group to be able to make additions and updates.  If you are publishing from ArcMap you will need to select Capabilities >> Feature Access to set these options.  These options can also be modified after publishing.

ThomasBjørnson_Larsen
Emerging Contributor

I only have access to the online tools. A Consultant have uploaded a featureset with about 100K features which runs perfectly fine on both pc and mobile. My question is here is if it possible to go beyond the 1000 limit within the Python API?

Also, is it possible to set the Create, Delete, Query, Sync and Update operations in Arcgis online or via the api?

0 Kudos
AndrewChapkowski
Esri Regular Contributor

Assuming you are using a spatial dataframe, you can publish the dataset as follows:

content.import_data(df=sdf, title='testabc123', tags='tags1,tags2')

0 Kudos
ThomasBjørnson_Larsen
Emerging Contributor

So, using spatial dataframe gets around the limit?

Do you have an example on how I can convert a pandas dataframe to a spatial dataframe?

0 Kudos